Senior Machine Learning Engineer, Learner Modeling

About the role

At Instructure, we believe in the power of people to grow and succeed throughout their lives. Our goal is to amplify that power by creating intuitive products that simplify learning and personal development, facilitate meaningful relationships, and inspire people to go further in their education and careers. We do this by giving smart, creative, passionate people opportunities to create awesome. And that's where you come in:

Our AI team is where a lot of that gets built: applying advanced AI to real problems in learning, and turning research into product capabilities that educators and students use every day.

We're looking for a Senior Machine Learning Engineer to build and own the learner models behind our mastery and progression capabilities. You'll design the models, build the pipelines that train and score them, and own their quality once they're running in production.

You'll partner with our learning scientists on what these models should measure, and with our infrastructure team on deployment and operations.

WHAT YOU'LL DO

- Design and build learner models, including knowledge tracing and longitudinal approaches, that power mastery and progression features surfaced to learners and educators

- Shape the data foundation for learner modeling: define which signals matter, and build the datasets your models depend on

- Translate mastery and progression definitions into model targets and evaluation criteria, working with learning scientists and product partners

- Build estimation and scoring approaches that hold up on sparse, noisy, and evolving behavioral data

- Own your models in production: training and scoring pipelines, testing, versioning, and monitoring quality once they're live

- Explain model behavior, assumptions, and limitations clearly to product, engineering, and learning partners

WHAT YOU'LL NEED

- Six or more years in applied machine learning, machine learning engineering, or applied research, with ownership of models shipped into real products

- Depth in at least one of: sequence modeling, latent-variable or probabilistic modeling, temporal modeling, Bayesian methods, or calibration of model outputs, applied to data that changes over time

- Strong Python and production engineering skills: you write the pipelines that train and score your models, and you've shipped models that run on a schedule and serve predictions to real users

- Strong evaluation instincts around calibration, uncertainty, stability, fairness, interpretability, and validation strategy

IT WOULD BE A BONUS IF YOU HAD

- Experience with recommender systems, user-state modeling, or personalization at scale

- Experience with knowledge tracing, psychometrics, educational measurement, or adaptive learning systems

- Experience combining structured knowledge representations, such as skills, standards, or concept graphs, with learner models